ABSTRACT Conventional methods for adult and pediatric tuberculosis (TB) diagnosis and treatment monitoring rely heavily on time-consuming bacterial culture or unquantifiable DNA assays to detect the presence of small numbers of bacteria. Pediatric TB management is particularly difficult because current organism-based methods for diagnosis and treatment monitoring are resource-intensive, invasive and inadequately sensitive for children, who tend to have paucibacillary disease. Due their immunological immaturity, young children are at high risk of rapid disease TB progression, with significant morbidity and mortality. Assessment of their treatment response relies on subjective measures, limiting children's access to much needed shorter and safer TB regimens. More rapid and rigorous methods of TB diagnosis and treatment monitoring are thus critically needed for children. In order to address current limitations in pediatric TB management, we propose to develop a rapid diagnostic method, independent of mycobacterial isolation, that quantifies the low molecular-weight M. tuberculosis (Mtb) antigens (CFP-10 and ESAT-6) and the TB-associated host marker IP-10 in patient blood samples. ESAT-6 and CFP-10 are ideal biomarkers to detect active TB and its treatment response, and addition of IP-10 to this Mtb-specific proteomic pattern will provide an integrated profile of patients' pathophysiological changes during anti-TB treatment, while also improving diagnostic sensitivity. Our strategy utilizes hollow, energy-focusing TiO2 NanoShells functionalized with custom antibodies specific for Mtb CFP-10 and ESAT-6 and host IP-10 peptides with high-throughput mass spectrometry (MS) to increase diagnostic sensitivity and specificity. We evaluated an Mtb-antigen-only version of this NanoShell-MS platform (no anti-IP-10 NanoShells) with 292 adult and 102 pediatric patients and controls chosen from five highly relevant cohorts (adult, pediatric and latent TB, HIV/TB co-infection, and non-TB mycobacterial infection), from our global collaborators. NanoShell-MS sensitivity and specificity for active TB (bacteriologically confirmed and clinically diagnosed) were, respectively, 90.7% / 97.7% in adults and 88.2% / 100% in children, with 78.6% sensitivity in culture-negative pediatric TB patients, greatly exceeding the WHO's published 4-15% Xpert sensitivity in this cohort. NanoShell-MS results can be obtained within one hour of sample collection compared to 4-6 weeks for conventional culture, and detected decreases in blood Mtb antigen levels within four days of anti-TB treatment initiation. In this proposal we aim to 1) use NanoShell-MS profiling to develop a quantitative prediction model for active TB diagnosis in large, well-described prospective pediatric TB cohorts and 2) determine the utility of our approach for rapid evaluation of treatment efficacy. Our NanoShell-MS assay platform has the advantage that it uses accurate, high-throughput mass spectrometry, which has become a critical means for clinical diagnosis in many parts of the world, while integration of our NanoShell-MS platform with an easy-to-use ?miniaturized? (shoebox size) MS system which should also allow this approach to serve pediatric patients in resource-limited areas.